I have been fiddling with teh grounding of my early model MFA Luminesence preamp. I think I’ve improved the original, but am interested in ideas and suggestions. It’s a unique design (to my knowledge), and I’ve never seen anything written that quite applies to it.
The design:
Phono stage and line stage. Uses octal tubes (5691, 6SN7, etc) with tube regulated power supply plus added filament regulators.
Separate boxes for power supply and preamp.
The power supply has three transformers--one for B + and two separate filament transformers, each regulated.
The original design has the preamp circuits separated by channel, bussed to a preamp chassis ground. The phono stage is closest to the chassis ground. All this is connected by an umbilical system (actually two umbilicals) that connects filament system and (regulated DC) as well as B+ and ground wires to power supply box.
Right now, the power supply box has a single star ground plus a separate mains ground both to chassis. Everything now connects to the first star, i.e., the power supply caps, the regulator ground and the ground from the preamp box. (I’m not home now, but will try attaching the preamp ground to the regulator ground.
One of the problems is that the main power transformer is very noisy, a defect of these early models. Still by experimenting, I’ve quieted the hum, though I’m not sure I’ve hit on the best solutions.
Any suggestions or general principles for such a design will be appreciated.
The design:
Phono stage and line stage. Uses octal tubes (5691, 6SN7, etc) with tube regulated power supply plus added filament regulators.
Separate boxes for power supply and preamp.
The power supply has three transformers--one for B + and two separate filament transformers, each regulated.
The original design has the preamp circuits separated by channel, bussed to a preamp chassis ground. The phono stage is closest to the chassis ground. All this is connected by an umbilical system (actually two umbilicals) that connects filament system and (regulated DC) as well as B+ and ground wires to power supply box.
Right now, the power supply box has a single star ground plus a separate mains ground both to chassis. Everything now connects to the first star, i.e., the power supply caps, the regulator ground and the ground from the preamp box. (I’m not home now, but will try attaching the preamp ground to the regulator ground.
One of the problems is that the main power transformer is very noisy, a defect of these early models. Still by experimenting, I’ve quieted the hum, though I’m not sure I’ve hit on the best solutions.
Any suggestions or general principles for such a design will be appreciated.
Actually, I can simplify this. My problem is how to deal with umbilcals, in particular what to do with the ground connection.
1) Connect it to the single star ground?
2) To the ground of the power supply?
3) To its own chassis ground.
I plan to experiment, but I'm interested in any cautions, warnings, observations, etc.
1) Connect it to the single star ground?
2) To the ground of the power supply?
3) To its own chassis ground.
I plan to experiment, but I'm interested in any cautions, warnings, observations, etc.
I JUST, I mean JUST bought the same three chassis Lumi off of ebay for..$106.05!!!! HOLEY SMOKES! WHAT A DEAL!!! (The preamp sells for nearly $5kUS, in ANY condition) No-one recognised it, it was so badly presented. I knew what it was, from a glimpse at the very first photo..and I nearly fell out of my chair. It sat at $29.99 for three days, no views, no bids. I still can't believe it. I'm a huge Bruce Moore fan, and this in my "Dream Preamp". I'm still pinching myself, and will be sweating until it actually arrives.
I'll be modding the crap out of mine, when it gets here.
I'll be able to contribute then.
I'll be modding the crap out of mine, when it gets here.
I'll be able to contribute then.
Hi Roger,
I think you want to keep the umbilical ground on the power supply star ground. Technically the AC ground should also be connected to this point.
I've forgotten whether your lumi has a bridge or pair of rectifier diodes for the high voltage, but the rectifier negative return (or center tap) should go to the EL input filter cap first and then from there on to the star ground, this will keep high cap charging currents out of the star ground connection.
I think you want to keep the umbilical ground on the power supply star ground. Technically the AC ground should also be connected to this point.
I've forgotten whether your lumi has a bridge or pair of rectifier diodes for the high voltage, but the rectifier negative return (or center tap) should go to the EL input filter cap first and then from there on to the star ground, this will keep high cap charging currents out of the star ground connection.
It's a closed auction, so you'll have to be a member, possibly, to look at it.
The chassis' are missing all three top covers, but they are going to do a good packing job for me, I paid them extra. The open chassis gives you a chance to take a decent look.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=9736790231&rd=1&sspagename=STRK:MEWA:IT&rd=1
At that price? I take a loan out and buy at least a few dozen.
That's obviously my pre-amp deal of the decade, and I've got nothing to complain about. It originally sold for $9 or 12k, I believe, depending on year. I think the three chassis was the last one, with very few examples made. As you say, it coulda been the earlier one, as the three chassis may have been too expensive to actually sell to unbelieving folks (with their mouths hanging open-stunned to silence-at the price) in the..what..early 90's, late 80's? I can't remember. It was certainly a contender for the finest tube preamp ever made. Unique in sound,as well. But I want neutral sound, not unique, so I'll be re-vamping where nessessary. No matter what, I'd be hard pressed to find any gear like this, that is so well laid out to - start modifying.
The chassis' are missing all three top covers, but they are going to do a good packing job for me, I paid them extra. The open chassis gives you a chance to take a decent look.
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=9736790231&rd=1&sspagename=STRK:MEWA:IT&rd=1
At that price? I take a loan out and buy at least a few dozen.
That's obviously my pre-amp deal of the decade, and I've got nothing to complain about. It originally sold for $9 or 12k, I believe, depending on year. I think the three chassis was the last one, with very few examples made. As you say, it coulda been the earlier one, as the three chassis may have been too expensive to actually sell to unbelieving folks (with their mouths hanging open-stunned to silence-at the price) in the..what..early 90's, late 80's? I can't remember. It was certainly a contender for the finest tube preamp ever made. Unique in sound,as well. But I want neutral sound, not unique, so I'll be re-vamping where nessessary. No matter what, I'd be hard pressed to find any gear like this, that is so well laid out to - start modifying.
yummy preamp
OK, the beast with three heads (chassis) has arrived...anyone got a schematic?
Or will I have to draw it out by hand? I'd hate to publish it, just to see some cheap *** boards show up on fleabay from the far east. Not fair to Bruce.
I guess either my Magus C or Couterpoint 3.0 will be for sale soon. Gotta keep one of them as a spare.
OK, the beast with three heads (chassis) has arrived...anyone got a schematic?
Or will I have to draw it out by hand? I'd hate to publish it, just to see some cheap *** boards show up on fleabay from the far east. Not fair to Bruce.
I guess either my Magus C or Couterpoint 3.0 will be for sale soon. Gotta keep one of them as a spare.
OK. One small problem. Damn, I'll come back in again.
Two small problems.
1) click/pop when switching volume level on switched ladder attenuator, but only in left channel, line or phono. Ladder attenuator only. Mostly lower settings. Not reproducable with other switches. Oops. I lied, just checked again. Reproducable with source switching, or level. Sounds a bit like RF or DC to me?
2) I noticed that the tonality is not linear, it depends on exactly what position the combination of left right gain (stepped, resistive) and stereo attenuator are in. I'll likely switch both to a shunt design to keep the loading even as possible. That should cure that weirdness.
The unit is DEAD QUIET. Zero hum, or buzz. This is interesting, as Bruce tends to put the gain control on the input of the line stage,...soo...this means the output section is so darned quiet..it sounds like silence when it's not doing anything. No mute output switch, which is alot of fun when you run fully DC capable 20 amp slo-blo fused power amps like I do. Usually the breaker pops... after the magic smoke gets out.
Gain is ridiculous. It is unusable, outside of having the -20db switch on.
I remember hearing that the Lumi would swing 100V p-p. I'm thinking that this might be true. (depending on loading, obviously)
And DAMN.. this preamp is dynamic. Whoa.
There is a set of RCA connectors on the back of the Phono section box, that says 'loading'. Presumably this is for phono loading. Is it strapped to 47k internally and then you load it down to where you want? Or does the unit need some 47k plugs to work properly?
Two small problems.
1) click/pop when switching volume level on switched ladder attenuator, but only in left channel, line or phono. Ladder attenuator only. Mostly lower settings. Not reproducable with other switches. Oops. I lied, just checked again. Reproducable with source switching, or level. Sounds a bit like RF or DC to me?
2) I noticed that the tonality is not linear, it depends on exactly what position the combination of left right gain (stepped, resistive) and stereo attenuator are in. I'll likely switch both to a shunt design to keep the loading even as possible. That should cure that weirdness.
The unit is DEAD QUIET. Zero hum, or buzz. This is interesting, as Bruce tends to put the gain control on the input of the line stage,...soo...this means the output section is so darned quiet..it sounds like silence when it's not doing anything. No mute output switch, which is alot of fun when you run fully DC capable 20 amp slo-blo fused power amps like I do. Usually the breaker pops... after the magic smoke gets out.
Gain is ridiculous. It is unusable, outside of having the -20db switch on.
I remember hearing that the Lumi would swing 100V p-p. I'm thinking that this might be true. (depending on loading, obviously)
And DAMN.. this preamp is dynamic. Whoa.
There is a set of RCA connectors on the back of the Phono section box, that says 'loading'. Presumably this is for phono loading. Is it strapped to 47k internally and then you load it down to where you want? Or does the unit need some 47k plugs to work properly?
-Linestage, and tape buffer. Tape buffer (it appears to be, I'll have to look closer) 12AU7. 6DN7 and 6SN7 in the line stage. Switched attenuator and switched left/right for balance. Separate left right signal switching, etc. Mirrored circuits.
-5691 and 5692 in the phono section, one whole box is phono, period. 6 octals. Mirrored circuits. Box is completely stuffed. Signal passing caps are about 16-20uf, per circuit group, in any given box.
-The bottom box is all PS, and about 40lbs of it. SS rectification. Mirrored circuits. Lotsa chokes.
Photos soon.
-5691 and 5692 in the phono section, one whole box is phono, period. 6 octals. Mirrored circuits. Box is completely stuffed. Signal passing caps are about 16-20uf, per circuit group, in any given box.
-The bottom box is all PS, and about 40lbs of it. SS rectification. Mirrored circuits. Lotsa chokes.
Photos soon.
Photos requested, photos supplied
This is a one-of-a-kind, pre-production hand-made unit, of the 'Venusian' series of Lumi-'ish' Preamps, from MFA. Perhaps Bruce made this one. I'll ask Scott Frankland when I go to get the schematic for it. Apparently, 'Randall Research' wire was all the rage at the time. I'm going to at least re-do the entire unit in silver teflon wire, to start........
There is only ONE of these. It will NEVER (over my dead body) be for sale. If you ever see it on Ebay or on Audiogon, the images have been stolen and the sale is BOGUS. Inform me, Audiogon, or Ebay..or come back to this thread and say so. Whatever the case, the images would be bogus and the sale a scam. There is only one of these. I own it, it's never going to be for sale.
Top down, Linestage images first. Hold onto your bandwidths! Here we go! Here come de gutty-wuts:
~~~~~~~~~~~~~~~
Most of the Left Channel (and some PS) for Linestage box:
Most of Right Channel (and some PS) for Linestage box:
Full Linestage box, from front above:
Appears to be the Tape buffer, One channel. Apparently...one of the great mods for these chassis is to remove the RF filtering. I wonder what could ever possibly make someone say that
:
More gutty wuts, from the back:
A closer look at the controls:
Front:
Back:
Bird's eye view:
And with the stainless lid on it, with damping - to control such a large plate of metal. This is why I suspect it came without the covers....they were removed as the unit sounds far better without them, if not propelry damped:
This is a one-of-a-kind, pre-production hand-made unit, of the 'Venusian' series of Lumi-'ish' Preamps, from MFA. Perhaps Bruce made this one. I'll ask Scott Frankland when I go to get the schematic for it. Apparently, 'Randall Research' wire was all the rage at the time. I'm going to at least re-do the entire unit in silver teflon wire, to start........
There is only ONE of these. It will NEVER (over my dead body) be for sale. If you ever see it on Ebay or on Audiogon, the images have been stolen and the sale is BOGUS. Inform me, Audiogon, or Ebay..or come back to this thread and say so. Whatever the case, the images would be bogus and the sale a scam. There is only one of these. I own it, it's never going to be for sale.
Top down, Linestage images first. Hold onto your bandwidths! Here we go! Here come de gutty-wuts:
~~~~~~~~~~~~~~~
Most of the Left Channel (and some PS) for Linestage box:
An externally hosted image should be here but it was not working when we last tested it.
Most of Right Channel (and some PS) for Linestage box:
An externally hosted image should be here but it was not working when we last tested it.
Full Linestage box, from front above:
An externally hosted image should be here but it was not working when we last tested it.
Appears to be the Tape buffer, One channel. Apparently...one of the great mods for these chassis is to remove the RF filtering. I wonder what could ever possibly make someone say that
:An externally hosted image should be here but it was not working when we last tested it.
More gutty wuts, from the back:
An externally hosted image should be here but it was not working when we last tested it.
A closer look at the controls:
An externally hosted image should be here but it was not working when we last tested it.
Front:
An externally hosted image should be here but it was not working when we last tested it.
Back:
An externally hosted image should be here but it was not working when we last tested it.
Bird's eye view:
An externally hosted image should be here but it was not working when we last tested it.
And with the stainless lid on it, with damping - to control such a large plate of metal. This is why I suspect it came without the covers....they were removed as the unit sounds far better without them, if not propelry damped:
An externally hosted image should be here but it was not working when we last tested it.
Excessive Tube Porn
An externally hosted image should be here but it was not working when we last tested it.
Hi KBK,
Both pairs of 5691 are in parallel in order to reduce the noise contribution of the input stage, this should be good for about a 6dB reduction in the noise floor relative to a single tube. (gaussian noise) The other side of the coin is that this topology has four times the miller capacitance of a single tube.
The Lumi seems to work best with fairly high output MC cartridges which are usually fairly insensitive to load capacitance. Certain Grado and Shure cartridges work well with low capacitance cable.
If this thing is anything like the Lumi that derived from it, it has feedback phono equalization and the ability to drive pretty low impedances as well.
Did Scott Frankland indicate how old this unit is? I would hazard a guess that it is in its late teens now at least and will need a lot of new electrolytics pretty soon.
I have worked on a number of Lumis in the past and Roger (Trombone) I am sure would be willing to share some anecdotes as he has owned one since the early 1990's.
As an aside I would not remove the emi/rfi caps, usually these are connected from audio input grounds to chassis ground and aren't in the audio path at all - they are however quite effective in making sure that the ground at the rca input jack is at rf ground. I doubt you can hear the caps on the supplies either as they are generally pretty well decoupled in this design.
Disconnecting the independent level controls on the left and right channels does seem to net some improvement in resolution.
I see many of what appear to be metal oxide resistors in these chassis, and REL Caps, many of these would be candidates for upgrading. Resistors are problematic because the dissipation in the cathode followers in particular is quite formidable. I would recommend VHCaps or REL TFT as I know these sound good in the LUMI at least. In the old days I used Holco, Vishay, and Roedersteins to upgrade resistors. The 5691 are a good choice sonically in this phono stage. I understand there are various improvements which substitute the 6EM7/6DN7 with other types (5692/6SN7 IIRC) with minor circuit mods and which SF indicates results in improved sonics.
Also talk to Trombone about mods as I seem to recall either Scott Frankland or Bruce Moore provided him with a few other worthwhile mods.
The power supplies in Trombone's unit (modded) were designed by me, but I no longer have any of the details, he might be able to provide some however.
Scott Frankland is a personable and helpful fellow in my few (not recent) dealings with him.
Both pairs of 5691 are in parallel in order to reduce the noise contribution of the input stage, this should be good for about a 6dB reduction in the noise floor relative to a single tube. (gaussian noise) The other side of the coin is that this topology has four times the miller capacitance of a single tube.
The Lumi seems to work best with fairly high output MC cartridges which are usually fairly insensitive to load capacitance. Certain Grado and Shure cartridges work well with low capacitance cable.
If this thing is anything like the Lumi that derived from it, it has feedback phono equalization and the ability to drive pretty low impedances as well.
Did Scott Frankland indicate how old this unit is? I would hazard a guess that it is in its late teens now at least and will need a lot of new electrolytics pretty soon.
I have worked on a number of Lumis in the past and Roger (Trombone) I am sure would be willing to share some anecdotes as he has owned one since the early 1990's.
As an aside I would not remove the emi/rfi caps, usually these are connected from audio input grounds to chassis ground and aren't in the audio path at all - they are however quite effective in making sure that the ground at the rca input jack is at rf ground. I doubt you can hear the caps on the supplies either as they are generally pretty well decoupled in this design.
Disconnecting the independent level controls on the left and right channels does seem to net some improvement in resolution.
I see many of what appear to be metal oxide resistors in these chassis, and REL Caps, many of these would be candidates for upgrading. Resistors are problematic because the dissipation in the cathode followers in particular is quite formidable. I would recommend VHCaps or REL TFT as I know these sound good in the LUMI at least. In the old days I used Holco, Vishay, and Roedersteins to upgrade resistors. The 5691 are a good choice sonically in this phono stage. I understand there are various improvements which substitute the 6EM7/6DN7 with other types (5692/6SN7 IIRC) with minor circuit mods and which SF indicates results in improved sonics.
Also talk to Trombone about mods as I seem to recall either Scott Frankland or Bruce Moore provided him with a few other worthwhile mods.
The power supplies in Trombone's unit (modded) were designed by me, but I no longer have any of the details, he might be able to provide some however.
Scott Frankland is a personable and helpful fellow in my few (not recent) dealings with him.
I just bought 4 Westinghouse NOS 6DN7's, and then another 8 that are identically branded (as a group), off ebay (all I want is two matched!). Tough to find those tubes. I'll sail along in stock tube compliment for the moment, and go after the caps, wiring, isolation and resistors first. Likely I'll do some minor circuit mods at the same time, like the removal of the gain controllers and switching to a shunt volume, if at all possible. The design is pre-lumi, ie a proto design example. The national red base 5691 tubes are insanely matched, a good thing, obviously. Very strong, too.
This is the 'original', that one-of-a-kind very first hand wired unit. The one that would be the 'workhorse'/ design example, is what I garnered from Scott's short issuance (noted here in this thread), and a phone call.
Due to it's nature, it is problematic to mess with, in terms of affecting value. Value is not my main concern though. Having a superior sounding preamp-is.
Already, it is the best sounding preamp I've ever managed to own and use. I'm looking forward to modding it. I'm likely to try and go full film caps on all B+ supply rails and all electrolytics in the garbage. Then local heater resevoirs right at the pins for that extra DB or so of dynamic expression..etc..etc. Right off the deep end. A good preamp to do it with. If you haven't tried that one before, do it last (Heater resevoirs at the socket pin), after everything else is done, so the situation is clear enough to hear it. The thing the double blind testers forget: that you can't hear a mod, if the rest is covered in crap.
To paraphrase Sun Tzu, 'wait long enough by the river and the body of your enemy WILL float by". In this case, I waited and waited and waited..and boom! my chance came at a ridiculously low price. Not exactly my enemy, by any measure, but the happenstance was fortuitous in a similar manner.
Right now, though.. the Mod Monkey on my back is scratching all of it's irritable bits on the Precision Fidelity M-8 power amp I bought about 1.5 years ago. It is responding nicely. I just bought 145 0.5uf 250VDC polystyrene caps. I keep finding uses for them.....
This is the 'original', that one-of-a-kind very first hand wired unit. The one that would be the 'workhorse'/ design example, is what I garnered from Scott's short issuance (noted here in this thread), and a phone call.
Due to it's nature, it is problematic to mess with, in terms of affecting value. Value is not my main concern though. Having a superior sounding preamp-is.
Already, it is the best sounding preamp I've ever managed to own and use. I'm looking forward to modding it. I'm likely to try and go full film caps on all B+ supply rails and all electrolytics in the garbage. Then local heater resevoirs right at the pins for that extra DB or so of dynamic expression..etc..etc. Right off the deep end. A good preamp to do it with. If you haven't tried that one before, do it last (Heater resevoirs at the socket pin), after everything else is done, so the situation is clear enough to hear it. The thing the double blind testers forget: that you can't hear a mod, if the rest is covered in crap.
To paraphrase Sun Tzu, 'wait long enough by the river and the body of your enemy WILL float by". In this case, I waited and waited and waited..and boom! my chance came at a ridiculously low price. Not exactly my enemy, by any measure, but the happenstance was fortuitous in a similar manner.
Right now, though.. the Mod Monkey on my back is scratching all of it's irritable bits on the Precision Fidelity M-8 power amp I bought about 1.5 years ago. It is responding nicely. I just bought 145 0.5uf 250VDC polystyrene caps. I keep finding uses for them.....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.